[PATCH] D39912: AMDGPU/SI: Implement d16 support for image intrinsics

[Changpeng Fang via Phabricator via llvm-commits]

cfang updated this revision to Diff 126799.
cfang added a comment.

1. Add definitions for new machine instructions for https://reviews.llvm.org/D16 bit and for whether target support packed/unpacked f16.
2. Add a flag bit to indicate whether an instruction is referencing 16-bit half typed data (https://reviews.llvm.org/D16).
3. Update to the LIT test cases to avoid using specific registers.

NOTE: We may define more instructions than necessary for the "PACKED", but I would appreciate suggestions  to get rid of them.


https://reviews.llvm.org/D39912

Files:
  lib/Target/AMDGPU/AMDGPUISelLowering.cpp
  lib/Target/AMDGPU/AMDGPUISelLowering.h
  lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp
  lib/Target/AMDGPU/InstPrinter/AMDGPUInstPrinter.cpp
  lib/Target/AMDGPU/InstPrinter/AMDGPUInstPrinter.h
  lib/Target/AMDGPU/MIMGInstructions.td
  lib/Target/AMDGPU/SIDefines.h
  lib/Target/AMDGPU/SIISelLowering.cpp
  lib/Target/AMDGPU/SIInstrFormats.td
  lib/Target/AMDGPU/SIInstrInfo.h
  lib/Target/AMDGPU/SIInstrInfo.td
  test/CodeGen/AMDGPU/llvm.amdgcn.image.d16.ll
  test/CodeGen/AMDGPU/llvm.amdgcn.image.gather4.d16.ll
  test/CodeGen/AMDGPU/llvm.amdgcn.image.sample.d16.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D39912.126799.patch
Type: text/x-patch
Size: 104323 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171213/eb5b5ff2/attachment.bin>